Open jcoffland opened 10 years ago
I'm working on a 2.5D path optimizer and looks promising. It uses two steps: firs it organizes the gcode into paths the second rewrites in order of nearest path firs comparing entry X Y values. The Gcode was generated using MakerCam, the preview images UniversalGcodeSender
Your method can give large improvements. I've implemented a more complicated and CPU intensive method that can give a bit better results. It uses simulated annealing. In some case it can be much better than the greedy algorithm. I've written the optimization code but have not yet finished it to the point where it can spit the GCode back out. See src/camotics/opt/.
Anyway your method is probably the preferred one in most cases simply because it's a lot faster to run.
Currently the tool path optimizer is useless because it can only read gcode and optimize the path but cannot regenerate optimized gcode.
See #9.